Cloudflare LAX Errors
No internet-wide outage is in progress this morning (as of ~09:50 UTC, Monday October 12). A handful of regional and single-provider incidents are open, and none of them look global.
**Cloudflare: elevated errors in Los Angeles (ongoing).** Starting at 08:52 UTC, Cloudflare says customers whose traffic reaches its Los Angeles (LAX) data center may see more errors than usual. The incident is still marked "investigating". In practice, sites behind Cloudflare may intermittently fail to load for some users on the US West Coast while traffic is mitigated or rerouted. Separately, Cloudflare's "Elevated Errors in China Network Colos" incident, first opened on October 10, was updated at 04:23 UTC today. Cloudflare describes slow speeds and intermittent timeouts between parts of mainland China and origin servers outside the region, across multiple ISPs, and says the congestion is upstream of its own network. Several individual Cloudflare locations are also listed as re-routed or degraded, which is routine for a network of its size. Source: https://www.cloudflarestatus.com
**Twilio: carrier-specific messaging delays.** Twilio opened several new incidents overnight: SMS delivery delays to multiple networks in Italy, SMS delays and some voice call failures to Pelephone in Israel, and phone number purchase failures in the new Console. Most of its other open incidents are long-running, carrier-specific SMS/MMS delays (Brazil, El Salvador, New Zealand, China, Australia). These affect message delivery on particular routes, not Twilio's platform as a whole. Source: https://status.twilio.com
**Minor, ongoing:** Akamai is still monitoring an APAC origin-connectivity incident that has been open since September 22. Anthropic is monitoring elevated errors on the Claude Console (platform.claude.com), open since October 7, and the Claude API itself is not marked as affected. OpenAI lists delayed cost data in its Compliance API only. Slack is still rolling out a fix for an iOS app crash when opening channels (since October 9). Azure DevOps shows a maintenance advisory in Australia.
**All clear:** AWS, Google Cloud, Azure, GitHub, Discord, Atlassian, Zoom, Vercel, Heroku, DigitalOcean, Cursor, Groq, ElevenLabs, Cohere and Stability AI show no active incidents. We couldn't get a machine-readable status for Stripe or Meta (metastatus.com) this run, and no outage reports for either turned up in search.
**Recently resolved (past ~72–96 hours):** Google Cloud Storage had elevated write latency and errors in us-central1 for about 12.5 hours on October 8 (preliminary incident report published: https://status.cloud.google.com). Cloudflare resolved several short incidents between October 7 and 9: API issues, a 1.1.1.1/Gateway DNS problem in Singapore, a major network performance issue in Columbus, and congestion in Querétaro, Sofia and Kochi.
**Crowd reports:** Over the weekend, user-report trackers showed brief spikes for some consumer services, including Imgur, Epic Games, Xbox Live, Overwatch 2 and the Japanese ISP J:COM. None of these were tied to a shared infrastructure cause, and we have no provider confirmation for them.
Overall: the internet is up. The Cloudflare LAX incident is the one to watch for West Coast users this morning.